
by Neruva
Neruva is a memory layer for coding agents, delivered as an MCP server backed by a hosted API. Agents recall earlier decisions, log what they do and leave handoff notes between sessions.
PluginContact for pricing
Step 1: Get an API key and a namespace from the maker.
Step 2: Install the @neruva/mcp server and configure it with the key and namespace in Claude Code, Claude Desktop or another MCP client.
Step 3: Start each session with agent_recall to load earlier decisions and mistakes.
Step 4: Use agent_remember and records_ingest to log decisions as they happen.
Step 5: End the session by writing a session_end record so the next session starts from it.
The pages describe Claude Code, Claude Desktop or any MCP client.
The client side is public on npm and PyPI. The server and its search and graph engines are not public.
Yes. Records can be forgotten by kind, tag or time, and an entity can be removed from the graph in both directions.
Requirements
CodeIndex MCPcodeindex is an MCP server that indexes a codebase with tree-sitter and answers structural questions such as symbol definitions, callers and change impact. It exposes 16 tools to AI coding agents.Repo2Graph MCPRepo2Graph MCP gives coding agents cited, token-bounded answers about a codebase. It indexes calls, imports and git co-change links, then returns the source blocks that answer a question.
CodeCartographer MCPCodeCartographer turns a repository into layered architecture notes, behavioral contracts, defect findings and a language-agnostic reimplementation spec. It ships as a Pi extension, an MCP server and a drop-in template.